过滤非汉字的utf8的字符

原文:http://www.oschina.net/code/snippet_564772_13507

utf8是变长字符集,单个字符占用1~4个字节。mysql在选择utf8字符集时,最多只能存储3个字节的 utf8字符,如果想要保存任意的utf8字符,数据必须用utf8mb4字符集,有些情况下,不能变更已选定的字符集,只好不得以而为之,把输入中的4 个字节的utf8字符全部过滤掉,好在,utf8字符集中,汉字是3个字节的。

public static String filterOffUtf8Mb4_2(String text) throws UnsupportedEncodingException {
byte[] bytes = text.getBytes("utf-8");
ByteBuffer buffer = ByteBuffer.allocate(bytes.length);
int i = 0;
while (i < bytes.length) {
short b = bytes[i];
if (b > 0) {
buffer.put(bytes[i++]);
continue;
}

b += 256; //去掉符号位

if (((b >> 5) ^ 0x06) == 0) {
buffer.put(bytes, i, 2);
i += 2;
System.out.println("2");
} else if (((b >> 4) ^ 0x0E) == 0) {
System.out.println("3");
buffer.put(bytes, i, 3);
i += 3;
} else if (((b >> 3) ^ 0x1E) == 0) {
i += 4;
System.out.println("4");
} else if (((b >> 2) ^ 0xBE) == 0) {
i += 5;
System.out.println("5");
} else {
i += 6;
System.out.println("6");
}
}
buffer.flip();
return new String(buffer.array(), "utf-8");
}
static public String filterOffUtf8Mb4(String text) throws UnsupportedEncodingException {
        byte[] bytes = text.getBytes("utf-8");
        ByteBuffer buffer = ByteBuffer.allocate(bytes.length);
        int i = 0;
        while (i < bytes.length) {
            short b = bytes[i];
            if (b > 0) {
                buffer.put(bytes[i++]);
                continue;
            }
            b += 256;
            if ((b ^ 0xC0) >> 4 == 0) {
                buffer.put(bytes, i, 2);
                i += 2;
            }
            else if ((b ^ 0xE0) >> 4 == 0) {
                buffer.put(bytes, i, 3);
                i += 3;
            }
            else if ((b ^ 0xF0) >> 4 == 0) {
                i += 4;
            }
        }
        buffer.flip();
        return new String(buffer.array(), "utf-8");
    }